The ability to externally control the expression of selected genes and thereby their gene products in field-grown plants by the application of appropriate chemical substances in the field can provide important agronomic and foodstuff benefits. This control is especially desirable for the regulation of genes that might be placed into transgenic plants and has many applications including (1) prolonging or extending the accumulation of desirable nutritional food reserve in seeds, roots, or tubers, (2) producing and accumulating products in plant tissues at a defined time in the developmental cycle such that these products are convenient for harvest and/or isolation, and (3) initiating the expression of a pest-specific toxin at the site of pathogen attack. The latter example may provide a means of avoiding contamination of the ultimate food product with the toxic agent as well as minimizing the development of resistance in the pest population by selective, tissue specific, rather than constitutive expression of the toxic agent. These and other benefits have been unattainable to date since a practical means to bring known plant genes under external control in the field has not been available.
In eukaryotic systems, the expression of genes is directed by a region of DNA called the promoter. In general, the promoter is considered to be that portion of DNA in a gene upstream from the coding region that contains the site for the initiation of transcription. The promoter region also comprises other elements that act to regulate gene expression. These include the "TATA box" at approximately 30 bp (-30) 5' relative to the transcription start site and often a "CAAT box" at -75 bp. Other regulatory elements that may be present in the promoter are those that affect gene expression in response to environmental stimuli, such as light, nutrient availability, heat, anaerobioisis, the presence of heavy metals, and so forth. Other DNA sequences contained within the promoter may affect the developmental timing or tissue specificity of gene expression. In addition, enhancer-like sequences that act to increase overall expression of nearby genes in a manner that is independent of position or orientation have been described in a number of eukaryotic systems. Homologs of these enhancer-like sequences have been described for plants as well. The vast diversity of promoter function in eukaryotic systems therefore provides the opportunity to isolate promoters with relatively stringent requirements for their transcriptional activation which may be useful in regulating the timely expression of gene products in transgenic plants.
While current technology exists to transform plants with the genes encoding selected products, the expression of these genes is either continuous throughout the life cycle (controlled by a constitutive promoter), or regulated by the developmentally timed program of maturation inherent in each organ/tissue/cell (stage or tissue specific promoters) in which the gene product is destined to be expressed. Continuous expression precludes controlled production of a gene product at particular stages of the life cycle, in specific tissues or in response to environmentally unpredictable events. In addition, such constitutive expression could place a major penalty on yield, due to greatly increased energy demands accompanying prolonged high level synthesis of a single gene product. Tissue or stage specific expression, although valuable for the temporal and spatial accumulation of products, is under the variable timing of the developmental program of each plant. The practical use of promoters from these types of genes would therefore necessitate the isolation of a multitude of stage-and tissue-specific promoters for all crop species of interest.
Ideally, one would prefer to externally control the expression of a gene product in transgenic plants by application of an inducing signal that stimulates expression of the desired gene in any tissue(s) at any time in the plant's life cycle. This regulation would be accomplished by controlling the expression of a structural gene encoding the desired product with a promoter that is highly responsive to application of the inducing signal. The proposed inducer/promoter combination should be functional in a wide variety of plant species, with the inducer having no effect on the normal plant growth, development or morphology. Chemicals that fit the above criteria for regulating gene expression in plants would be of great utility in the field, as their use would be compatible with current agricultural practices. For instance, application of a chemical inducer could be easily accomplished using equipment currently in use by most plant growers. Ideally, a chemical/chemically responsive promoter combination could be made functional at any stage or in any tissue of a transformable plant to control the expression of any desired gene product.
There are inducer/promoter combinations that have been shown to regulate the expression of foreign genes in both bacterial and animal systems. Many of the inducible bacterial systems are based on the use of promoters that respond to metabolites or metabolite analogs that normally regulate bacterial growth. Addition of an appropriate metabolite to the media of active growing bacterial cultures transformed with genes driven by promoters that are responsive to these metabolites results in expression of the desired product. Examples of such inducer/promoter combinations include 3-.beta.-indoylacrylic acid/Tip promoter, IPTG/lac promoter, phosphate/phosphate starvation inducible promoter, and L-arabinose/ara B promoter combinations. Similarly, heavy metal/metallothionine promoter, and heat/heat shock promoter combinations have been used in animal cell culture systems to control the expression of foreign genes.
There are a number of inducer/promoter combinations derived from plant genes that are known. Activation of many of these promoters is regulated by environmental factors such as light, heat shock and anaerobioisis. The promoters of these inducible genes have been extensively analyzed [c.f., Kuhlemeier et al., Ann. Rev. Plant Physiol., 38:221-257 (1987)]. However, the use of environmental inducers for regulating foreign genes is impractical since the inducing signal (i.e., light, temperature and O.sub.2 levels) are not easily or practically controllable under conditions of normal agronomic practices. Other plant genes have been described that are induced by oligosaccharides, such as those generated during pathogen infection and/or wounding. Examples include the induction of phenylalanine ammonia lyase and chalcone synthase by glucan elicitors in soybean [Ebel, J., et al., Arch. Biochem. Biophys. 232, 240-248 1984] and induction of a wound-inducible inhibitor gene in potato [Cleveland, T. E. et al., Plant Mol. Biol. 8, 199-208 1987]. Again, the promoters of these inducible genes lack utility in regulating the expression of foreign genes in transformed plants due to either lack of a practical method of induction (wounding) or the deleterious effects that result from diverting metabolic energy from plant growth to large scale synthesis of products designed to combat pathogen attack (oligosaccharide inducers).
A large number of chemicals, both natural products and synthetic compounds, have potential use in controlling gene expression in plants. However, any chemical that may be useful as an inducer of gene expression in the field must minimally be environmentally safe, have little or no effect on the normal growth, morphology and development of plants, and be easily used under conditions of normal agronomic practice.
A number of natural products are known that affect gene expression. These are mainly naturally occurring plant growth regulators such as the auxins, cytokinins, gibberellic acid, ethylene and abscisic acid [c.f., Davies, P. (Ed.) Plant Hormones and Their Roles In Plant Growth and Development, Martinus Nijhoff Publ. 1987], while other chemicals have equally dramatic effects such as salicylic acid [Hooft Vanhuijsduijnen et al., J. Gen. Virol., 67:235-2143 1986]. When the growth regulators described above are applied to various plants or plant derived cells/tissues/organs, a change in the metabolism is observed that has been shown to be due, at least in part, to new gene expression. Some products of these genes as well as the genes themselves have been isolated and characterized. However, since the chemicals that induce these genes normally function in regulating the growth and development of plants, they cannot be candidates for inducers of recombinant, chemically inducible genes in transgenic plants. This lack of utility is a direct result of undesirable pleiotropic effects that would arise from the undesired co-activation of the plant's endogenous hormone sensitive developmental programs along with the desired recombinant gene. For example, activation of a foreign gene by abscisic acid in developing plants would induce many undesirable hormone effects including negative effects on plant metabolism [Milborrow, B. V. An Rev. Plant Physiol. 25, 259-307 1974], a sharp decline in growth rate, an induction of stomatal closure, and premature abscission of young leaves and fruits. Other phytohormones have similar negative effects on plant growth and development that preclude their use in regulating the expression of foreign genes in transformed plants. A more general review of phytohormone effects on vegetative plants including ABA, ethylene, cytokinins, and auxins, is presented in Phytohormones and Related Compounds: A Comprehensive Treatise Vols I and II, Letham, D. S., Goodwin, P. S., and Higgins, T. G. V. eds. Elsevier/North Holland (1978).
Among the potentially attractive chemical candidates that may have utility in regulating gene expression in transgenic plants is the group of compounds collectively called herbicide antidotes or safeners. Safeners are functionally defined as chemicals that have the ability to increase the tolerance of a crop plant to the toxic effects of herbicides when the plant is treated with the safener. It now appears that the safening action of these compounds is related to their ability to increase the metabolism of the herbicide in safener-treated plants [Sweetser, P. B., Proceedings of the 1985 British Crop Protection Society Conference-Weeds. 3:1147-1153 1985]. For example, treatment of maize and other cereal crops with safeners such as the dichloroacetamides increases their tolerance toward several groups of herbicides [Lay, M. M., and Casida, J. E. Pest. Biochem. Physiol. 6:442-456 1976, Parker, C. Pesticide Science 14:533-536 1983]. More specifically, N,N-diallyl-2,2-dichloroacetamide safening is correlated with an increased level of glutathione-S-transferases (GSTs), a family of enzymes known to detoxify several major classes of pre-emergent, selective herbicides by conjugating them with glutathione [Mozer et al., Biochemistry 22:1068-1072 1983]. This increase in GST activity is correlated with an increased steady-state level of GST mRNA in treated plants, as shown by the work of Wiegand et al [Wiegand, R. et al., Plant Mol. Biol., 7:235-243 1986]. Thus safener treatment of selected plants can increase the steady state level of a gene product without having significant effects on growth and morphology.
It has been shown that changes in the rate of metabolic detoxification of sulfonylurea herbicides in corn plants are induced by treatment with a variety of safeners [Sweetser, P. B., Proceedings of the 1985 British Crop Protection Society Conference, Weeds 3:1147-1153 1985]. The result of this accelerated metabolic detoxification is increased herbicide tolerance in safener-treated plants. For example, 2 to 5 fold increases in the metabolism rates of chlorsulfuron and metsulfuron methyl have been observed in wheat and corn following application of the antidotes napthalic anhydride, N,N-diallyl-2,2-dichloroacetamide, or cyometrinil. This observed increase in sulfonylurea herbicide metabolism occurs within hours following antidote treatment. In addition, the safening activity of the chemicals is not seen if plants are treated with the protein synthesis inhibitor cycloheximide prior to safener treatment, indicating that the increase in herbicide metabolism is dependant on de novo protein synthesis. This requirement for new protein synthesis indicates that safener treatment may activate the transcription of specific nuclear genes, and that a safener/safener-induced gene promoter combination may exist that will have utility in regulating the expression of foreign genes introduced into transgenic plants. To date, however, there has been no reported example of an inducible expression system for transgenic plants based on activation of safener-responsive promoter/structural gene recombinant DNA construction by the external application of a safener or safener like compound. Indeed, no system with real utility for externally regulating the expression of a desired gene in transgenic plants that is compatible with current agronomic practices is known.
The instant invention focuses on DNA promoter fragments derived from several plant species which are inducible by herbicide safeners of cereal crops These promoters have been used to develop a safener/safener inducible gene system for controlling the expression of foreign genes in transformed plants. This system has utility for externally regulating the expression of desired genes in transgenic plants in a grower's field. Its advantages include the high level of activity shown by several of these promoters in response to application of an appropriate inducing chemical, the apparent expression of these promoters in all plant tissues tested to date, and the absence of pleiotropic effects generated by treatment of plants with these chemicals.
Ebert et al., [Ebert et al., Proc. Natl. Acad. Sci. (USA) 84:5745-5749 1987], discloses studies of the active fragment of DNA containing the nopaline synthase promoter. This promoter is constitutive rather than inducible, and while of bacterial origin, operates in a wide range of plant tissues. A construction was made so that the promoter controlled the expression of the reporter gene chloramphenicol acetyl transferase (CAT). The authors reported that a fragment of 33 bp (-97 to -130) of DNA was sufficient to promote expression of the CAT gene. They reported further that the presence of two copies of the fragment tripled the expression of the CAT gene. These results from stably transformed tobacco tissue were repeatable in a transient assay using tobacco protoplasts. Comparison of the level of CAT activity obtained when gene expression was controlled by the 33 bp fragment in both the transient expression and stably transformed tobacco protoplasts and tissues resulted in some differences. The authors nevertheless indicated their belief that such transient assays are valuable for studies of promoter sequences in stable transformation systems. Operable linkage of the nopaline synthase promoter to a structural gene, however results in constitutive expression of the gene product in transformed plants precluding its use in externally controlling gene expression.
Studies of the anaerobic induction of the maize alcohol dehydrogenase (Adh I) gene by electroporating gene fragments of Adhl into maize protoplasts from suspension culture cells have been performed [Howard, et al., Planta, 170:535-540 (1987]. Transformed protoplasts were subjected to reduced oxygen levels and assayed for Adhl expression 20 hours later. To facilitate measurement of anaerobiosis-induced Adhl gene expression, the 5' promoter or regulatory fragment of the native Adhl, gene (1096 base pairs) was functionally linked to a CAT gene. Their results demonstrated the normal anearobic regulation of the inducible Adhl promoter/CAT gene from a monocot maize gene (i.e., Adhl) in protoplasts derived from a homologous cell culture system. They also showed that the Adhl promoter fragment, without the coding and 3' regions of the Adhl gene, is sufficient for anaerobic induction of a foreign coding region in maize protoplasts.
Other researchers [Lee et al., Plant Physiology 85:327-330 1987], have further defined the size of the DNA fragment responsible for anaerobic induction of the maize Adhl gene. These researchers transformed maize protoplasts with a recombinant gene consisting of a CAT coding region under the control of the Adhl promoter and measured the production of CAT 24 hours later. By modifying the length of the promoter fragment used in the construction, Lee et al. determined that 146 bp 5' to the transcription start site were sufficient to place the expression of CAT under anaerobic induction. However, the expression of CAT was increased 5.times.or 8.times.by the addition of 266 or 955 bp, respectively, of contiguous 5' promoter sequences.
Walker et al., [Walker et al., Proc. Natl. Acad. Sci. (USA) 84:6624-6628 1987], continued the studies of the DNA sequences in the promoter region of the maize Adhl gene required for anaerobically induced gene expression in a transient assay. They determined that control of anaerobic induction of gene expression resided in two sequences from the promoter: those being the sequence between -133 and -124 bp and the sequence between -113 and -99 by (5' to the transcription start site). Both sequences are necessary for induction. Attachment of the full 40 bp element to an unrelated viral promoter conferred anaerobic regulation to the chimeric promoter.
Others have shown that extremely low levels of CAT gene expression could be observed under appropriate anaerobic conditions when the DNA fragment between base pairs -1094 and +106 bp of the maize Adhl gene was used to regulate CAT gene expression in stably transformed tobacco cells, [Ellis et al., EMBO Journal 6:11-16 1987]. In fact, only CAT messenger RNA was detected. However, promoter elements from the octopine synthase gene of bacteria, or those from the Cauliflower Mosaic Virus (CaMV) linked 5' to the Adhl promoter, stimulated the expression of the CAT gene and permitted detection of CAT after anaerobic induction. The fragment of DNA consisting of 247 bp obtained adjacent and 5' to the transcription start site of the structural gene for Adhl, was sufficient to put the expression of the CAT gene under anaerobic control. Therefore, anaerobic control by the 247 bp fragment of DNA was maintained even when the octopine synthase and CaMV 35S promoters, which are constitutive promoters, were present. The region of the Adhl promoter responsible for anaerobic induction demonstrated in transient assays by Howard et al., Lee et al., and Walker et al. were similar and identical to the region showing anaerobic induction in stably transformed plants by Ellis et al.
Patents have been issued to animal and microbial systems in which the expression of selected gene sequences have been induced by chemicals that interact with certain regulatory sequences. U.S. Pat. No. 4,579,821 issued to Palmiter and Brinster discloses the isolation of promoter/regulator sequences of the mouse metallothionein-I gene and its use to control the expression of selected DNA sequences operably linked to the promoter by exposure to heavy metal ions or steroid hormones. The expression of thymidine kinase fused to the metallothionein-I promoter was obtained in differentiated cells of adult mice upon administration of cadmium or dexamethasone. U.S. Pat. No. 4,703,005 issued to Nakata and Shinagaua discloses the isolation of a gene for phosphate-binding protein (phoS) to which was fused a foreign gene 3' to phoS. The foreign gene is controlled by phosphate in the culture medium. None of these inventions, though has any potential utility for use with plants in the field. The heavy metal ions that activate the metallothionein promoter are both toxic to plants and would pose an extreme environmental hazard in the field. Similarly, promoters responsive to nutrients such as phosphate lack utility due to the requirement of plants for constant levels of these nutrients for normal growth in the field.
Several reports of attempts to regulate the expression of genes in transgenic plants have been reported. European patent application number 85302593.0 discloses the isolation of four heat shock gene promoters from soybean and claims their use for driving the expression of foreign genes in transgenic plants. In the application, the authors claim the use of these promoters in temporarily activating expression of foreign genes such as a crystalline toxic protein structural gene of Bacillus thuringensis or an herbicide resistance gene in response to heat stress in vivo. However, this leaves the expression of a gene linked to one of these heat shock promoters to chance changes of the daily temperature in the field.
Marcotte and Quatrano [J. Cellular Biochem. Supplement 12C, 1988; Marcotte, W. R., Bayley, C. C., and Quatrano, R. S., Nature 335, 454-457 (1988)] have reported initial results of studies of the inducibility of a chimeric gene whose transcription is driven by promoter fragments derived from two abscisic acid (ABA)-inducible genes (Em and a 7S globulin) from wheat. The products of these genes were shown to be induced in whole plants by addition of ABA. The induction was shown to be, at least in part, at the level of transcription. Promoter fragments of varying lengths from the 5' region an Em genomic clone were translationally fused to a bacterial .beta.-glucuronidase (GUS) coding region that was linked to polyadenylation signals from the CaMV 35S transcript. The ABA inducibility of GUS activity using these different length promoter fragments was analyzed in transient expression assays using both monocot (rice) and dicot (tobacco) protoplasts. They demonstrated that regions upstream of the Em coding region (650 bp) and the 7S globulin coding region (1800 bp) contain sequences that are sufficient for ABA-regulated expression of GUS activity in rice protoplasts transient assays. The Em promoter failed to show any responsiveness in the dicot transient expression assay, indicating that the promoter may not function in dicot plant species. However, as discussed in detail in an earlier section of this work, the induction of undesirable pleiotropic effects resulting from application of phytohormones (including ABA) to whole plants in the field precludes the use of these compounds in regulating gene expression in transformed plants.
A patent was issued in Europe to De Danske Sukkerfab A/B [CC87-106623] that claims a method to improve the nitrogen fixing system of leguminous plants by controlling the expression of genes of interest with a promoter from a root/nodule specific gene. Specifically, the inventors demonstrated that a chloramphenicol acetyltransferase (CAT) gene driven by the promoter derived from a soybean leghemoglobin gene was inducible in the roots of transformed plants in a fashion similar to other root specific genes that are affected by nodulation. The method is severely limited in that induction of genes is limited to simulation by nodulation and the induction is root specific. It cannot provide a true means to externally control the expression of genes at any time in all tissues of field grown transformed plants.
To date, there are no reports of practical means to externally regulate the expression of foreign genes in transgenic plants using a method compatible with those used in normal agronomic practices. While reports of plant promoter sequences stimulated by light, heat, anaerobic stress, and phytohormones have appeared, no disclosures of specific inducible promoters that are responsive to chemical substances that might constitute the basis for a practical method to control gene expression in plants by application of the chemical in the field have appeared. At this time, a clear need exists for such promoter sequences to be used in recombinant DNA constructions that would enable one to externally control the expression of genes that can confer agronomic advantages if expressed at the proper time. Further, this specificity of expression should be amenable to external control through exposure of plants to chemical substances which can be readily applied by a variety of application methods and which only induce the expression of the desired target gene.